After holding with black in game three, Anand is back looking for his first lead of the match. Official site.
Update: Wow, Anand just devastated Topalov with a knight sac and mating attack out of another Catalan. Yes, a knight sac and mating attack on move 25 out of a Catalan. Topalov may have to try an Indian Defense against the Indian if this keeps up. The world champion takes his first lead of the match. A really beautiful combination, with Black finding it impossible to prevent a white rook from lifting to the 4th rank, either Rc4 or Rd4, and delivering mate. Those watching the video said that Topalov offered a very limp handshake upon resigning on move 32, but to be fair it's hard to reach across the board when you've just had your head shoved up your rectum like that. Takes a lot of yoga.
What a match, with 3/4 games decisive to start! Topalov played the routine 22..Rad8 very quickly, allowing the sac on h6. I'm sure computers will show very quickly if that was the losing blunder or if he could have saved himself after the sac. Looks very hard to do, so right now we can consider 22..Rad8 a catastrophic lapse in Topalov's sense of danger. Once again he shows himself to be an impatient defender against slow pressure. I wonder if Anand has Kramnik stashed in that van he drove from Frankfurt? 22..Nc5, heading to d3, runs into 23.Rc4! with similar threats to the game. 22..Rfd8 23.Ne5 Be8 is disgusting, as is 22..f6. But defending the Catalan well usually requires a level of passive groveling and that's just not in Topalov's DNA, bless him.
Anand leads 2.5-1.5 going into the rest day on Thursday.
Bonus analysis from mishanp in the comments: "Just for the record - a 2:2 draw on toilet visits today."
R4: Anand played 15. d5. Was it not better to 15. axb - cxb - 16. d5- ??
I am sure he is still in his preparation.
No, the immediate d5 is better in these positions, because it threatens to undermine the defense of the b-pawn, and bxa4 is never a serious concern. If the pawn trade happens first, then there is no threat to make the b-pawn vulnerable, so the c4-pawn gets to breathe easily.
Hotep,
Maliq
Slightly OT - not sure if the chat applet was a good idea. Don't like what it's doing right now...
Now 16.dxc6 would be vintage Vishy ! Has he still got it ?!
Harish Srinivasan & Maliq >
dxc6 will be played if he was playing Kramnik. But I think he may not do it here. But then I want him to play it.
So Anand is now playing as Kramnik again - Topolov tries to stir things up (and he's had a while since the Elista match to think about it) - and finds himself on the rough end of Anand's preparation. If Anand can win the other day with a slow crush - surely today its Anand with the initiative again.
d5 is a thematic break in many openings like QGA (Isolated D pawn in QGA) . This is to open up the position for the beautifully centralized pieces. Black's pieces lack co-ordination.
White will not want to resolve the tension on the Q-side unless he has tactical/strategic justification. axb5 only plays to black's advantage since he has to resolve the tension there an emerge with the consolidated material advantage after the exchange .
i think he will play ng4.. why he is taking so much time for that
i want anand to win.. but i like topas style of playing
i was right.
is anand still in his preparation..?
You should know. You share your name with her wife.
I don't think there is a preparation more. too many variations he has to count...
Correction.
I mean -
You should know. You share your name with
HIS
wife.
Your comment "chat if you've got it" Mig reminded me of Slim Harpo's "I've Got Love If You Want It".
http://www.youtube.com/watch?v=5JLtWvUbZvM
all of the sudden Anand switch gear to the king side with Ng4 .... is he trying to be Topalov in this game? :-)
Anand knows how to confuse Topalov. 22. Ng4, is just kidding to sacrifice it on h6, but he will go on e5. Anand doesn't risk.
Anand must be thinking hard about Nxh6 now .... (me too)
OOops, he did it. Typic!
Vishy's back !! 23.Nxh6+ !
he did it !!! Nxh6. a lone Queen on the king side for the moment, need to swing one of rook over there quick by pushing on e5
Well that was a precise analysis, mansoor! Maybe Anand does not read Daily Dirt...?
This will be such a brilliancy if Anand pulls it off.
Can he really do that?! I suppose the point is that e5 comes with tempo on c6 and then you needn't recapture on g2 because Rc4 is so quick. Is that right? Anyone with a computer?
Maybe Bh3?
Wow. First Topalov invites to exactly this kind of thing with ..h6, then he "ignores" (allows) it almost without even thinking... How nonchalant.
25 e5 bxge 25 pxf6 followed by Qg6+ Kh8 then Rc4.
Don't know if that will hold ... there is a reason why I am not playing in WCC :-)
No, that was nonsense, wasn't it - 25 e5 Bxg2 26 Rc4 Bf3, for example. Perhaps his idea is 26 exf6 as roamingwind says.
Maybe I should do some work instead.
Are we all too proud to have computers going then?
Yayyy 25. e5 played!
Vishy finds the silicon move 25.e5!!
at this point chess.com evaluates to "decisive advantage for White", given 25...Qg7.
advantage anand.. but topalov still has good chance
@rdh: but if 26. Rc4 Bf3 27. Rh4 ?
Okay, so this will be the first time since Elista that Topalov falls behind in a match. Brace yourselves..
Incredible shot by Anand. Topy looks crushed on Rybka. This sac now shows Anand playing like Topalov. Haha.
Go Vishy
yes anand wins today. 2.5-1.5
now the question in my head is why the heck did Topalov moved "20 ... h6" ? there is nothing going on on the king side then, and there is no back rank weakness on black side (so that h6 would allows the king to escape in case of rook check on 8th rank).
topalov in time trouble
Yes I didn't understand 20..h6, but it doesn't have to be bad in itself. But you HAVE to spend more than 4 minutes (!) after Ng4 before you decide if Nxh6+ works or not.. that was how long it took him to play 22..Rad8.
Man, that was a royal whupping. The Champ bares his teeth!! Topalov's anti-Catalan play is being blown apart, he loses the thread in the middlegame...play the Semi-Slav, man!!!
Anand will push hard and quick so Topalov does not have enough time to think.
27...Qg7 then, I suppose.
He was preparing Ra8 to d8.
This looks like some Alekhine game from the 1920s against a minor master.
Way to go, Vishy!
@why the heck did Topalov moved "20 ... h6" ?
not wise enough to think that after seeing his 2 knights stuck on the q-side Anand with switch to attack on the k-side ( amusingly there is such an advice and examples given in A. Kotov in the (famous) book "Think like a grandmaster")
Anand completely outplayed Topalov in this one. Very impressive.
D.
agree. h6 is not bad it itself. also agree that Topalov somehow completely missed Nxh6. very unlike him ... given that sacrifice for attack is his style.
Anand completely outplayed Topalov in this one. Very impressive.
D.
I think black played 27 ...Be4 to block the white rook on c1 from swinging to the king side via c4.
Susan Polgar says: Qg5+ and following by Rxe6 Anand wins. But I don't understand how? After Nd3 !! from Topalov...
Anand still has to follow up with correct moves here. it is not quite over yet.
Oh, I found it! After Nd3 Anand will play Rc5!
Oh NOoooooooo! Rc5! was a winning move but Anand played Rc2?
I think Anand should offer a draw now!
i think qf5 was not a good move
And he is wining...
He has won anyway!
it's all over... white now has the classic rook on the seventh with queen on the h file and black king is all exposed.
Might that knight on a6 not enter the play strongly after Rc5, mansoor?!
1-0!!!!!
1-0. Anand has pleasantly surprised me with his nerves of steel, after losing Game 1. Two consecutive white wins in such contrasting styles!
Heavy psychological blow to get such a crushing!
If antics are to play any more of a role in this match we might well see them soon...
y topalov resigned?
Sure, it was maybe longer way to win. I allready analysed it and short and save way is already played. tnx.
And Topalov resigns! Congrats to Anand on an perfectly played game (unless Rybka finds something unexpected), Topa must be shellshocked...
Just two brief quotes from Short before I go:
"This has been a brilliant game by Anand. First he used great skill to obtain an edge but the tactical punishment of the careless h6? has been brutal."
"This has been superb stuff from Anand - far better than that streaky win in the second game."
Vishy seems to be at the peak of his powers!
Today, Anand played the opening like Kramnik2006, but the middlegame like Anand (or Kramnik2009?). Topalov played like ... ?? Danailov will play the postmortem like ... ??
Manu sure is quiet today.
Of course it was a streaky win after Short called one of the key moves "shockingly bad", and said White had "no winning chances whatsoever".
On 20 ...h6, Polgar's analysis was:
An unexpected but useful move. Black now has the option of getting his a Rook to d8 in the future without worrying about back rank checkmate problem.
Let's see if she changes it retrospectively, as usual :)
Kapalik
Ok, and one last not so brief before I have to run :)
"That was a superb attacking game from the World Champion - Anand at his absolute best. Breathtaking chess. After playing like my grandmother in the first game, he has come back with a vengeance. His win in the second game was a bit scratchy but this jaw-dropping stuff - a delicate positional sacrifice to begin with followed by sudden unexpected unbridled aggression!"
The Champ proved who's the best.
Ng4 pays off
Vishy is following Garry's advice after game 2, that he can just "give up a pawn in every game".
The attack quite came quite unexpectedly since it looked like all the play was on the queen side. And once Topalov Queen went to a7, suddenly from no where the attack came on king side. Simply brilliant and Topalov was completely outplayed... although GM analysis might reveal that h6 and alter Rad8 were the losing move.
h6 was too bold to be true... having the knight discover the queen. I like it if moves like h6 are punished well.
Mig and lot of other players are saying that Anand is Kramnikian, I am not sure whether that is a correct way to describe his style. i understand that he is playing a particular style which kramnik has used all his life successfully, but Anand showed how to deal with it. So Anand is just as everyone calls more practical and plays as the "position demands" as he is saying often now a days.
Maybe shorter matches encourage a greater willingness to play for three results?
And I wonder - how many players in history have as much/more natural chess talent as Anand? That list has got to be really short.
Interesting musing by Mig, might be on to something there...
Perhaps Kramnik was more than willing to offer Vishy help in this match. After all, there is no love lost b/w Topalov and Kramnik, and by seconding Vishy, he could get an "inside" view of Vishy's preparations, which is certainly what helped him defeat Kasparov in 2000 (Kramnik seconded Kasparov in 1995 versus... Vishy!).
Interesting to think about, anyway.
You can bet on it. Unless she reads this blog.
I'm not calling Anand Kramnikian, I'm calling his play with white in this match Kramniknian. Just like his play with both colors, especially black, was much sharper, even Topalovian, against Kramnik in their match. He's showing tremendous versatility and was quick to switch to terminator mode when he had the chance today.
Mig sez: "I wonder if Anand has Kramnik stashed in that van he drove from Frankfurt?"
Methinks, great as Kramnik is, that is giving WAY too much credit to the stem game, which deviated as early as move 10. Come on, this is vintage Anand, who is a wonderful positional player but also a devastating tactician. I think he definitely edges Kramnik in that area.. We have a Champion worthy of all his great predecessors.
I was at work, briefly glanced at the opening, and then could only check back after the game was over and I saw Armageddon had been unleashed!!
The key here is that Vishy is in form and seems to be prepped to his eyeballs, his G1 blunder notwithstanding. However don't write Topy off yet, its just -1. Long way to go folks!
Anand is playing like Kramnik and Topa is playing like Susan Polgar. A lawsuit will be filed by the end of the match.
I am not a chessgames Premium member, so I have to rely on someone in the Chessbomb chat quoting Short earlier in the game. From my memory, apparently he said something like "Anand again blundered a pawn in the Catalan, now he will exchange queens and win the game".
Hammer time!
Anand can play like Kramnik to beat Topa, or he can play like Topa to beat Topa. What's a guy to do?
But don't read too much into one or two games. "Topa can't play positional chess" or "he has no impulse control" are childish simplicities. Please take away my impulse control right now and make me a 2800 player. He WILL be back.
Nah, Anand is playing like Anand.
What do you all expect in the next game ? e4
Now that his beloved Topa-boy has fallen behind, perhaps it's time for Danailov to come up with some nasty stuff. Hope Vishy and team are prepared for anything and everything.
Ok, ok. Anand isc playing like Topa and Topa is playing like VanWely. Topa will finish a respectable second (-3) in this event.
Mig
Please let us have any comments from Gary K if possible. Would love to hear his thoughts on the match as well...:)
It could be that playing at home and all the weight of expectations is bogging T down.
Sorry... "Gary" = "Garry"... no offence intended
@mig.. "Topalov may have to try an Indian Defense against the Indian if this keeps up...defending the Catalan well usually requires a level of passive groveling and that's just not in Topalov's DNA, bless him."
If hasn't concluded yet (after so many unplesant experiences) that he simply can't play (for long) this kind of positions what on earth will convince him now ? He should have drawn these conclusions already after Elista.
Anand not doing so well. He will finish next to last.
So folks think that Anand is Kramnikian or he is Topalovian!
So it looks like Anand does not have an identity of his own!
Which means he is nothing on his own!
What a judgement of a champion player!
And not to forget that more than any other player Anand seemed to always have to prove he is good to belong at this level. Kramnik or Topalov never had to do that presumably!
Now that it seems like Anand has a terrific match strategy (with the exception of the Game 1 hiccup), Ovidiu's comment strikes me as very apt. Did he just not expect this kind of play? Did he go into this match with his weaknesses very clear or is Anand just excellent at exploiting them?
Hope so! LOL
Updated MC simulation probabilities.
Probability distribution prior to BLITZ and after SPEED.
Topa win: 23%
Anand win: 73%
Draw: 4%
Dramatic change in prob distribution after game #4!
Now, Topa needs to win 5 out of 8 in order to win within the classical round - slim chance (22.5%)
It should be obvious to Topalov by now that his ability to defend against the Catalan is just not up to par, especially against the likes of Anand. From a match strategy perspective, I'm really not sure why Topalov would even allow himself to be thrown into positions with Black which are essentially passive and thus psychologically uncomfortable for his aggressive and impatient nature. I expect to see a Queen's Indian from Topalov next time ... or maybe even a Benoni??!!
Trust Ovidiu to find a way to defend Topalov and somehow bring an unique insight to a crappy comment.
Ovidiu, your Topa can neither "passively grovel" nor can he "actively stand up" :-)
What are we to do with him then :-)
In many competitions, Topa has been behind many times only to recover and win. The probabilities (see above) don't take into account the personality. Topa has this history of catching up....
"Now, Topa needs to win 5 out of 8 in order to win within the classical round - slim chance "
I guess you mean 5 points out of 8 (not 5 wins). He only needs two wins to take the lead.
If Anand draws the next and somehow wins with white...its match over for Topa..
Correct! Five points. Thanks.
@ From a match strategy perspective, I'm really not sure why Topalov would even allow himself to be thrown into positions with Black which are essentially passive..
Vanity perhaps, he doesn't want to accept that he can not.
Why Kasparov allowed the Berlin-wall over again, or Kramnik played twice into Anand's home prepared sharp Bb7-Meran ?
Though I am rooting for Vishy, the match is far from over. Kudos to Topa for playing double-edged stuff as black and risking losses such as the one today.
Watching the live video feed is awesome! Vishy hesitated before playing the brilliant e5--he put his hand out to push the pawn, then withdrew his hand and thought some more.
In his excellent write up, Mig reports some viewers' comments about Topa's resignation handshake. Topa simply resigned courteously and professionally. What's he supposed to do, jump over the table and kiss Anand?
Willie The wonka has revealed Topa's brilliant strategy for the match: by forcing Anand to win game after game, Topa ensures that the Indian GM will finish a miserable next to last, while Topa will be a very respectable second!
Once again, Danailov shows the world what a truly unique strategist he can be. It's a pity he let his own chess career go down the toilet!
Willie The wonka has revealed Topa's brilliant strategy for the match: by forcing Anand to win game after game, Topa ensures that the Indian GM will finish a miserable next to last, while Topa will be a very respectable second!
Once again, Danailov shows the world what a truly unique strategist he can be. It's a pity he let his own chess career go down the toilet!
Bob, a match is very much different compared to a round-robin tournament. There is no bottom-feeder to exploit here, no opponent who suddenly has to adjust from meeting Leko to witnessing firebrand chess. Furthermore, in a tournament, the possibility that one will never again in the same event be posed with the problems that plagued him in a previous game (due to opening preferences of opponents) is much different than in a match, where we can reasonably expect that the victor in the previous game will again question the loser in a similar variation. As Topalov learned in Elista (where he had to resort to shenanigans in order not to lose outright in the classical games), match play is an entirely different animal.
Hotep,
Maliq
I don't know about 25 e5 being brilliant. It's hard to imagine any other move.
Incidentally, Shipov also gave 20...h6 a ! in real time. I doubt it's such an obvious mistake as Anand made it look. As to 22...Rad8, I'd be more inclined to mark it a terrible blunder if I could see what else Black's supposed to do.
I don't understand all the Ovidiu hate on this board. Perhaps other threads in the past have set the tone, but nothing here seems particularly biased. In fact, he seems sort of tough on Topalov in this thread.
Folks should read the comment before they read the author...
That's actually not a bad point, Ovidiu! ;)
One of Anand's great merits so far is not sticking to the Gruenfeld after game 1.
> So folks think that Anand is Kramnikian or he is Topalovian!
> So it looks like Anand does not have an identity of his own!
> Which means he is nothing on his own!
> What a judgement of a champion player!
In martial arts that is considered the very highest level of fighting skill, where one just follows the opponent and situation. to have a style is to have a target for exploit. As Kasparov once said (paraphrased) "they can change their openings but they can not change their styles." (in referring on how he played over his opponents' games to discover their styles/weaknesses)
I loved the game by Anand. Clearly Anand has learnt and adapted over the years how to channel his talents into a winning strategy. When he played Kasparov in 1995 he did not really have psychological strategy only over the board preparation and strategy.
In his matches with Kramnik and now with Topalov he is demonstrating how truly universal his play is and how he continues to learn and adapt.
For someone who can be categorized as being mainly an E4 player, to use D4 for the most of the two matches so far at older age speaks volumes about his talent and drive to make himself better.
For me personally that has been most remarkable insight into who he is as a player in the past 2 to three years.
This match is only 4 games old. Here is wishing for some more exciting on the board chess moves.
Mig, thanks for commenting on my thoughts, but I may be wrong in logic here, but first thought is such statement mean that Kramnik is good and Anand is not in such positions, if at all I will say they are both equal. Obviously Anand can play good positionally is what you are suggesting, granted then people will say finding Nxh6 is more like Topalov. I am sure for Anand whatever is analysis up till and until I win thats good. Related to this what people think is one of the best Anandian game if there is one, of course it need not be Anand's.
Rdh, I suggested Bh3 earlier, because although intuition called for e5, I couldn't tell how to continue after Bxg2, since recapturing was obviously losing. Of course, Anand was able to see through it clearly, which is why he is 700 points stronger than I am! For what it is worth, even though e5 was the intuitive move, it still should be regarded as a powerful follow-up, especially given the fact that it was mandating the sacrifice of a second piece (since allowing the Bc6 to be lost for nothing would not be a serious attempt at weathering the storm and testing the soundness of the sacrifice).
Hotep,
Maliq
Agreed. Subtleties can't be easily modeled through simulations. The assumption I made is that the games outcomes are independent (from game to game). Clearly, this is too much of a simplification.
In reality, because of your considerations, the probability of Topa winning might be even slimmer than 23% (see above). But then again ..... Topa does not easily give up. When he is behind he tends to focus better - can't model this either without resorting to all kind of distributions (Prob(Topa) catches up when behind).
Wow. It's good to see a World Champ so dominant. In this game Anand combined the best of Kramnik and Aronian: Solid structure out of the opening looking for minuscule advantages (the Kramnik part), then suddenly some "strange" moves leading to a fatal blow (the Aronian part).
As much as one would enjoy Topalov's dynamic play and opening novelties, it's impossible not to admit Anand's superiority (so far).
Great match!
rdh, agree. That is why this game will be remembered for a long time. At no point there is a blunder, yet topa gets destroyed. Anyone can claim any move before nxh6 a blunder post hoc, but they were not really in a live game.
Agree with e5 also, it was between e5 and bh3, and other's analysis indicates that both would have won.
I don't recall Topalov being beaten so soundly in a long time, both strategically and tactically. Vichy drew his attention (and everybody else's, I suspect) towards a positional repetition of G.2, but then slashed it short. Topalov's team have some thinking to do now. The roles are completely reversed now to what it was after G1. This match may be thriller, but for now Topalov has to even reach there cause in the last 3 game she seems like out of it..
D.
I believe that Anand's wins both result from outplaying Topalov over the board, while Topalov's win comes from home preparation, and possibly a memory lapse by Anand. If that's all true, it bodes well for an Anand match victory. As it should - I think we'd all like the better "player" rather than the better "preparer" to win.
You make too many assumptions, all of them against Topa. One thing for sure though: The better man will win!
@bob: Would you please give some details on the setup of your MC simulations. Are the probabilities based on match position? How are they set ? Surely not based on ELO.
what is MC simulations?
Monte Carlo simulations, ref "http://en.wikipedia.org/wiki/Monte_Carlo_method#Monte_Carlo_simulation_versus_.E2.80.9Cwhat_if.E2.80.9D_scenarios".
URL got cut off for being too long, I guess. You can visit "http://en.wikipedia.org/wiki/Monte_Carlo_method".
MC stands for Monte-Carlo. Comes from Stats/Probability domain. Allows you to generate many possible combinations of moves
Dan, the simulation is based on probabilities updated after each game. I look at the history and estimate the probabilities for "win" "draw", "lose" for a SINGLE classical and rapid game. Then I simulate the match for given current score.
For example, in this case I used the following probabilities:
Classical: Topa(win, draw, lose) = (0.25,0.5,0.25)
Rapid: Topa(win, draw, lose) = (0.03,0.64,0.33)
The formula is:
Prob(Topa_Overall_Win) = Prob(Topa_win_prob_classic) + Prob(draw_match_classic)*Prob(Topa_win_prob_rapid)
The assumption is that the match will not reach BLITZ phase.
Don't discount Topalov yet. This guy is at his most dangerous when he's wounded. He's come back to win many tournaments in which he had a poor start.
There's not really much to add from seeing that game live, except... it was a privilege to be there! Two wins and a draw for Vishy while I've been in Sofia - my work here is well and truly done :)
Today seemed to be the first game of the three when Vishy was really pressing in the opening, but even today Topalov put up tough resistance and made Vishy sit and put in some very deep thought in the middle game - but as in all the games the time was obviously well spent and it was Topalov struggling on time and with the position by the end (obviously a bit more spectacularly today!).
I actually thought Topalov had weathered the opening storm until he allowed dxc6 and after that Vishy was almost playing with him. He didn't play that quickly after the sacrifice, so even though black's forces were stranded away from the king and there where devastating ideas in the air it wasn't clear if Vishy had it all worked out. And then, as someone mentioned, he withdrew his hand the first time he went to play e5, just to add to the tension!! (I haven't checked any analysis yet, but Bh3 also looked interesting there)
I didn't think there was anything particularly odd about Topalov's resignation, though great write-up, Mig :) There was even less applause than after Vishy's last win, but on reflection I think it's mainly just that it's a small audience. I don't know how wide the frame of the live camera is, but there were moments that would have made great photos (if cameras were allowed) - with Vishy standing and calmly looking on while Topalov had his face buried in his hands, desperately trying to find salvation. Chess is a cruel game!
Just for the record - a 2:2 draw on toilet visits today, and no rogue alarms troubling the silence. There was a nice murmur around the hall when Anand sacrificed his knight.
Mig, any word from Kasparov if he was watching? He would have appreciated the sac!
@bob, a very easy-to-do improvement for including match position would be - the variability of result would increase i.e.
Topa(win, draw, lose) = (0.25,0.5,0.25) would change to (0.3,0.4,0.3) and if he is still down, would eventually go to (0.5,0,0.5). Though more likely to (0.45,0,0.55)
I will be curious to know of Nxh6 -- not in this position but in any position resulting from this opening came up in Anand's preparation. I doubt it. Main reason being, ...h6 seems to be a very non obvious move in the catalan since all play is on queen side. And with the amount of time spent in the middlegame by the players in this game, it certainly seemed that this was calculated over the board.
On the other hand if it had come up in his preparation, then Anand's preparation is extremely deep and thorough.
So, is it author's insinuation that Kramnik is Anand's secret second?
I had to work and couldn't follow the game online, but this thread makes a great read. I can understand what was going on and re-live the excitement. Much better than any computer-perfect objectivity.
In the Anand-Kramnik match Anand not only defeated Kramnik, but humbled him enough for Kramnik to think about his approach to chess and change it. I hope Topalov will get humbled, think and change, too. Maybe by the end of this match he can acknowledge a drawn position without arbiter assistance, and even follow up with a proper handshake.
Good, classic play by Vishy today - first get the upper hand by fine positional play, then start the tactics.
Bob, I do sociological research, so I am more used to predicting large-scale probabilities as opposed to outcomes for an individual case and have no confidence in my ability to statistically do so. (So, for example, I'd be more comfortable predicting that an "average" 2800 would have whatever probability of rallying in the match based on a number of predictors than arguing that Player X had this probability, given that there are so many unknowable factors.) Therefore, while I will say that match play statistically is a different beast compared to round-robins (and, in turn, round-robins compared to swiss events), I won't be one to count Topalov out at this point, especially given the fact that Anand has basically just "held serve" by winning both of his whites. Topalov is quite capable of beating any player in the world from either side of the board, and all he needs is a single victory for the outlook of the match to be entirely different again.
Hotep,
Maliq
poor audio press conf is up on official site
http://www.anand-topalov.com/en/gallery.html
Dan, the match position has been included in my simulation. The probability distribution (0.25,0.5,0.25) is based on previous history + results from the first 4 games. The current score (1.5: 2.5) is taken into consideration; thus, it has been accounted for that Topa needs 5 points from the remaining games to win the classical round.
This isn't a tournament it is a match. Unlike a tournament Topalov cannot draw games and have Anand lose games and gain ground. He must win games to gain ground, so just on that face alone it is tougher for him to make a comeback should he continue to lose.
"..So, for example, I'd be more comfortable predicting that an "average" 2800 would have whatever probability of rallying in the match based on a number of predictors than arguing that Player X had this probability, given that there are so many unknowable factors..."
------------------------------
Maliq, Any probability estimation is a subject to an error. Also, re ELO 2800 or so; should it be taken into account? Not sure because there is no transitivity in chess. For example A may win against B, B may win against C. Does it mean that A will win against C? Not really. Some chess players are very inconvenient to play, although their ELOs may not be that impressive. The transitivity does not hold in chess.
Personally, I believe that the personal history between the two players is more important than the ELOs.
:-)
These types of probablity models are too simplistic to have any kind of reliability. The whole idea of giving a pre-defined probability distribution like Topa(win, draw, lose) = (0.25,0.5,0.25) or (0.3,0.4,0.3) for all the remaining games is pretty much worhtless. That's because (0.25,0.5,0.25) may be true for game-5, but cannot be true for game-12 depending on the score at that point. If say, the score is Vishy-Top 6-5 after game 11, then the probability should be more like (0.45,0.05,0.50) or something similar. In other words, Topalov will go all out in such a situation changing the individual game's chance of ending in a draw pretty low. A good model needs to factor in these situations. At the very least the model needs to give different distribution for White and Black games.
Bob,
Yes, I know that all estimations are subject to error. My issue is with how to interpret the findings and make claims based on them; sociologists are notorious for avoiding the exceptions in favor of the rules, but if a given competitor is just such an exception, then this is clearly problematic when predicting the outcome of a match in which he is competing.
Also, do not other factors move the winning probability for a given game, as well? For example, it is generally acknowledged that Kramnik's Berlin defense diffused the probability of Kasparov winning with white greatly, and there is no real way to know how each player will proceed in future games, although it is reasonable to assume that Anand will continue to question Topalov's ability to meet the Catalan.
Hotep,
Maliq
Thank you and other guys, I learned quite an imp stuff today.
Just one thing to add on yoga... the only danger Vishy seemed to be in today was of falling head-first off his chair onto the board! He's got a way of precariously perching himself on the very edge of his seat & sitting bolt upright when he really wants to get into the position. Topalov's the opposite and ends up with his spine almost parallel to the board.
Juan, it can never be (0.45,0.05,0.50)after #11. The highest probability would still be "draw" because of the previous history, prior to the match. Besides, the MC simulation has been tested many times and very often it works pretty well even in situation more complex that this one. Besides, in the beginning of the match, the MC simulation resulted in an edge of Anand over Topa. So far, it works in the predicted direction.
RE the white and black thing - it would refine the numbers but would not dramatically change the results - in this match, the two players have equal # of whites and blacks. Only if the history is dramatically different from 50:50 whites/blacks, it would matter. It is very unlikely though; the two have played 48 times with 50:50 chance for whites/blacks. The law of the large numbers kicks in.
My prediction is: Anand wins
Besides, I am a Topa's fan.
@r: "Kudos to Topa for playing double-edged stuff as black and risking losses such as the one today"
Dude, "double-edged stuff" means a complicated position which has opportunities for both sides to play for a win. I fail to understand at which stage Topa had any opportunity to win!
You need to improve your chess vocabulary.
"...although it is reasonable to assume that Anand will continue to question Topalov's ability to meet the Catalan."
--------------
Fine, it will be reflected in the simulation (which is constantly updated after each game) because the probability distribution changes and temp current score will be correlated with this.
Bob, the (0.45,0.05,0.50) thingy after #11 shouldn't be taken literally. My point is that the individual game's probability distribution should be influenced by the match situation; and so, any model which fails to consider that is severely unreliable.
In any match between 2 players of close strength, whenever one goes ahead, his/her chance of winning the whole match increases significantly. In this case more so, because Anand is a superior rapid player which increases his chance even more. So, there's a high chance that your prediction will come true and Anand will prevail. But, the probablity you are assigning based on your model seems unreliable to me.
Congratulations to Mr. Anand!
Vishy has played calm, clear chess in a very difficult arena. Topalov is a formidable opponent; I admire his undisturbed approach in the face of adversity. Keep up the great work!
Congratulations to Mr. Anand!
Vishy has played calm, clear chess in a very difficult arena. Topalov is a formidable opponent; I admire Anand's undisturbed approach in the face of adversity. Keep up the great work!
G1: Topa: Nxf6!! +-
G4: Anand: Nxh6!! +-
-----------------
"Alekhine's attacks came suddenly, like destructive thunderstorms that erupted from a clear sky." -- Kasparov
for instance note Be4 in:
http://www.chessgames.com/perl/chessgame?gid=1012080
or Bf6 in:
http://www.chessgames.com/perl/chessgame?gid=1012053
Anand also seems to be a master of sudden attacks as can be seen from his win today and also several others like:
http://www.chessgames.com/perl/chessgame?gid=1563806
(Anand - Svidler, 2009)
http://www.chessgames.com/perl/chessgame?gid=1482300
(Anand - Topalov, 2008)
http://www.chessgames.com/perl/chessgame?gid=1452484
(Anand - Carlsen, 2007)
".. My point is that the individual game's probability distribution should be influenced by the match situation; and so, any model which fails to consider that is severely unreliable.
In any match between 2 players of close strength, whenever one goes ahead, his/her chance of winning the whole match increases significantly. In this case more so, because Anand is a superior rapid player which increases his chance .."
------------------------------
Juan, all of the above is part of the model!!! I thought that I described all this clearly in the above postings. Note that I am using different distributions each time a game is played; the rapid distribution also differs from the classical.
That's why Topa's chances went down dramatically after four games, being behind on Top of it. Also, his weakness in Rapid is another reason why he does not fair better.
I think if Topa unleashes a strong novelty in a rapid game, Anand might find it difficult to deal with it OTB given the time constraint.
I remember Carlsen too echoed the sentiments that Topa is not a formidable rapid player, it might be offset a bit by the fact that in WC he might be willing to unleash prepared moves which he otherwise would not.
Dana MacKenzie translates Sergei Shipov's coverage of R4: http://www.danamackenzie.com/blog/?p=811
Agree, a novelty during Rapid may be decisive! Also, Topa played Rapid [relatively] well against Kramnik.
"Alekhine's attacks came suddenly, like destructive thunderstorms that erupted from a clear sky." --
G. Kasparov
"Alekhine was never a hero of mine. His style worked for him, but it could scarcely work for anybody else. His conceptions were gigantic, full of outrageous and unprecedented ideas. It's hard to find mistakes in his games, but in a sense his whole method was a mistake."
R.J.Fischer
"Alekhine is a player I've never really understood; yet, strangely, if you've seen one Alekhine game you've seen them all. He always wanted a superior center; he maneuvered his pieces towards the King's-side, and around the twenty-fifth move began to mate his opponent"
R.J.Fischer
Without feeding in the moves and backtracking, I am curious to know which engine found that Nxh6 is winning for white and how long it took?
Fritz 9 and Rybka 2.3 dont find it after 5 min. d 14
just use this position
3r1rk1/q4pp1/n1bNp2p/p7/pn2P1N1/6P1/1P1Q1PBP/2RR2K1 w - - 0 23
interesting Harish...makes us think on how deep the preparation could be from Anand and his team.
In game 1 apparently Topalov did not play one move out of his prep.
"Alekhine is a player I've never really understood; yet, strangely, if you've seen one Alekhine game you've seen them all. He always wanted a superior center; he maneuvered his pieces towards the King's-side, and around the twenty-fifth move began to mate his opponent"
R.J.Fischer
!!
"I fail to understand at which stage Topa had any opportunity to win!"
Today, (as far as I can tell) Anand was never in danger of ending up worse - he played too well for that ... . But the whole concept of black trying to hold on to the extra pawn in the Catalan is double-edged: if white can neither regain the sacrificed pawn nor keep enough compensation, black will get winning chances.
BTW, this is also Kramnik's recent interpretation of the Catalan when he has black "playing against himself" - and he scored 2.5/3 against Gelfand (Tal Memorial 2009), Topalov (Zurich Champions rapid) and Carlsen (Corus 2010). He may have been worse at some stage in all of these games - "no risk no fun".
I understand that you are "using different distributions each time a game is played." But that is only based on the result of the last game to change the probability distribution.
Say, up until one particular game, Anand won x games, Topalov won y games and z games were drawn between tham. Correct me if I am wrong, your model calculates the prob of the next game as Anand(win, draw, lose) = (x/(x+y+z), z/(x+y+z), y/(x+y+z)). So, if Anand actually wins that game the prob changes as follows for the next game:
Anand(win, draw, lose) = (x+1/(x+y+z+1), z/(x+y+z+1), y/(x+y+z+1)).
What I am saying is, this model is too simplistic. Because it doesn't take into account the match situation.
Is there any other significant chess life left for a World title holder and defender except preparing for the championship match?
Bob: "Personally, I believe that the personal history between the two players is more important than the ELOs."
Of course, but there is so little data available for the two, that the Elo data for 2750+s is bound to be better.
Instead of (0.25,0.5,0.25), starting with something like (0.28, 0.60, 0.12) [with 0.28 for White] would make more sense. Then, gradually increase or decrease the drawing chances as the match situation changes.
Or, just go a simple fixed model of (W=0.28, D=0.60, B=0.12) for each game because, really, fretting about a few per mil or percent isn't worth it.
Juan,
From the given distribution, I simulate the outcome of the whole match, not the next game. That's why one needs MC; this is not about the next game. Again, so far, MC correctly caught where the match would go!
"Of course, but there is so little data available for the two, that the Elo data for 2750+s is bound to be better. "
---------------------
Historically, they played 48 times - that's not so little. In the beginning of the match, the previous history is used (44 games); then it is updated after each game. It seems to me better than USING ELO. If we had to use ELO only - then the match would go for Topa. Does not look that way, as of now.
I am always amused when the Electric Light Orchestra is mentioned in chess threads.
Elo is a distant second.
ooh, evil wo-man!
I think I found the answer to my own question. I just downloaded Firebird 1.2 (since frtiz and rybka 2.3 dint find Nxh6) and its supposed to be the strongest engine right now (as per http://www.chessvibes.com/reports/chess-engine-controversy/ )
And it takes full 4 min. at depth 16 to find that Nxh6 is winning. (ofcourse no feeding the move and backtracking). It took a genius to find it over the board (as i said earlier, i doubt it is home prep although chessvibes commentary thinks its possible )
Knallo,
Elo/ELO: That's all you have on the subject? Besides, ELO is the correct term for chess. Elo is a breed of dog. Here is some info:
"...The Elo is an emerging breed of dog, with development beginning in 1987 in Germany. The breed name is trademarked and development has been closely supervised by the Elo Breeding and..."
Let me know if you are fascinated by Elo dogs. Will give some more info.
I re-read my post and realized that it is improperly worded. Carlsen only indicated that Topa is not too strong a rapid player. The comment that he might play prepared moves is my own. I thought this correction is in order in the interest of "academic honesty" :).
chessdom said this as the game was going on
...
What about 23.Nxh6+ now ?? It certainly is a winning line 23. Nxh6+ Kh7 24. Ng4 f6 25. e5 Bxg2 26. Kxg2 fxe5 27. Kg1 e4 28. Qe2 Kg8 29. Qxe4 Qe7 30. Qe5 Rd7 31. Rd4 Nd5 32. Ne4 Nab4 33. Qh5 Nf6 34. Rxd7 Qxd7 35. Ngxf6+ gxf6 36. Qxa5 but will Anand find it? Stockfish at Chessdom / Chessbomb found it at high depth.
23 Nxh6 is not the most difficult move in this game by a long way; any top player could calculate to 26 exf6 very quickly. I'm amazed the engines don't go for it much more quickly; I suppose it must be because Black stays material up for a long time.
re: Elo/ELO
That's really funny, Knallo. Good point! I think I've always refered to "ELO ratings" myself.
Ah I see. thanks. Stockfish at Chessbomb might be running on very fast processors that might have enabled them to give that variation quicker than the time taken by firebird on my machine.
To verify, I did try downloading stockfish 1.6 and ran on my machine. It actually took only 3min. (a min. less than firebird) to find Nxh6
Anand played the move in less than 5 min. So only 3 possibilities
1. It was home prepared
2. It was an intuitive sac
3. He saw the follow up and calculated it all the way.
I go with the third choice.
Now I am going to run an engine match of stockfish vs firebird :)
Bob: actually, Knallo is right about ELO/Elo, though it's a mistake that we chess players often make. Arpad, the inventor of the rating system, had the last name Elo, not ELO.
Is source code available for any of these engines (need not be the strongest)?
@rdh "23 Nxh6 is not the most difficult move in this game by a long way; any top player could calculate to 26 exf6 very quickly. "
wow..."any top player" ..."very quickly"..... so easy to say.
Well you differ with my view. fine.
Firebird I think is open source, but I dont where to find it, although I obtained the exe. I know that crafty source is available and I have it.
@the inventor of the rating system, had the last name Elo, not ELO
yes but when we mention IT we refer to the system not to WHO invented, just as when we use 'newtons' for measuring forces, not Newtons
@rdh "@rdh "23 Nxh6 is not the most difficult move in this game by a long way; any top player could calculate to 26 exf6 very quickly. "
Oh sorry, I misunderstood you. You are probably saying Topalov is not a top player for otherwise he would have seen Nxh6 coming, because it was very easy to calculate not just that, but even upto exf6.
Well here too I disagree with you as I feel Topalov is a top player.
Yes, Ovidiu, and when we refer to the system, we should use the inventor's last name, Elo, not an acronym, ELO, which has nothing to do with chess ratings. Just Google it: many instances of "Elo ratings" (for football, too, not just chess!). I'm not sure if there are any but mistaken instances of "ELO ratings."
Anyway, this is all rather silly--we should be celebrating today's crushing victory by Anand.
I found this blog following Ian Rogers at USCF site links to some good sites for post match analysis. Very interesting read.
http://dejanbojkov.blogspot.com/
I agree with rdh that Nxh6 is not hard to find in that position, purely because even I considered it as soon as Anand put the knight on g4. Topalov must have noticed it, but for whatever reason he dismissed it (he played rad8 relatively quickly) ... don't think he is a bad player for that by any means ... just unlucky oversight.
Lots of blogs are asking what tricks SD may be about to pull if Topalov continues to lose - here is an interesting piece of "breaking" (sic) news from the official webpage:
"Breaking news:“The Father of the Euro” Prof. Mundell will arrive in Bulgaria at the personal invitation of Mr. Silvio Danailov", ...aha.... Eurogate....
@Oh sorry, I misunderstood you. You are probably saying ...
you don't understand well chess, Srinivasan, ..both of them saw, considered, 23.Nxh6 but Topalov misjudged it, underestimated it, after a superficial thinking. He played 22..Rad8 rather quickly convinced that he would be able to cope with Nxh6 if played
Bob: "Historically, they played 48 times - that's not so little."
In the past five years, they've played 15 games with classical time controls. That is very paltry data.
And a trinomial random variable has so little information that even if you take older games from a decade or so ago, n = 48 is still pathetically small, so any probabilities you define (e.g., (.25-.5-.25)) based on such a weak data set are bound to substantially miss the mark.
Bob: "If we had to use ELO only - then the match would go for Topa."
Huh?! You don't seem to have any concept of what Elo really is or measures. Topalov's rating is higher, so at the start of the match Elo would say, "Topalov has a greater chance of winning than does Anand," and not at all that "the match would go for Topa."
Bob: "Besides, ELO is the correct term for chess."
Ha!
Definitely not. But if you don't know that, I certainly will not be able to convince you that Arpad Elo did not consider his last name an acronym. Just keep it up, and me amused.
Wow, Topalov got pwned like the punk biatches he and his crew of thugs are. I hope Vishy wins every single game hence. I was ever a fan of trash talkers...
Any updates on a potential toilet-gate incident? 10 bucks for anybody who can come up with a potential gate candidate.
Deep Rybbie 3 finds Nf6+ after about 90 secs at depth 12. It took about four mins and depth 14 for Nh6 to climb to the top of the move tree. It first showed up at about 2:30,
This on a rather modest dual core machine. I am sure a good quad core will be much faster.
@ "Besides, ELO is the correct term for chess."
ELO (not Elo) writing is the most often seen, read, that's why it feels as "naturally" the correct one.
But if you want to be snobish you can have it as Elo and we will guess what you mean.
There are many incorrect things that are "often seen."
For instance, in chess movies, the lower-right-hand square is often shown as black. Clearly we should start setting up our boards that way, to avoid being snobbish.
You probably dint get what I was hinting at. I was just trying to point out that Nxh6 and seeing all the way to exf6 is far from easy since rdh was saying so. I dont understand what your point to me was ?
sample from the press
http://www.china.org.cn/sports/news/2008-07/01/content_15916808.htm
Here you go - Elo is a breed of dog and the name is protected.
-------------------
Dominguez, First Cuban to Surpass 2700 ELO Coefficient
Adjust font size: ZoomIn ZoomOut
Great Master (GM) Leinier Dominguez held Monday the best punctuation obtained by a Cuban in the world chess classification, accumulating 2,708 units of ELO coefficient (it is a system that measures the player's skills levels) .
In the new listings of the World Chess Federation (FIDE-International Federation of the Discipline), broadcasted this Monday, Dominguez sustained the 25th place and surpassed with 13 points accumulated from March to June, a national record until then (2,695).
This is Dominguez' sixth consecutive improvement and the fifth new mark for Cuba since he completed 2,678 points in April 2007 and ranked in the 30th position.
His excellent performance in the Capablanca Memorial in Havana, and the Sarajevo Magistral in Bosnia, has put him the first place in Latin America.
At continental level, Dominguez is only excelled by Gata Kamsky from the United States (27th place with 2,723 points).
The exclusive mark of 2,700 is only surpassed now by 29 GM, leaded by world champion Viswanathan Anand from India, whose ELO decreased from 2,803 to 2,798.
Anand is followed by Russian Alexander Morozevich (2,788) and Vladimir Kramnik (2,788), Ukrainian Vassily Ivanchuk (2,781), Bulgarian Vesselin Topalov (2,777) and Norwegian Magnus Carlsen (2,775).
The prestigious group of the best 10 chess players in the world is completed with Teimour Radjob (2,744) and Shakhriyar Mamedyarov (2,742) both from Azerbaijan, Alexei Shirov (2,741) from Spain and Peter Leko (2,741) from Hungary.
(Xinhua News Agency July 1, 2008)
And, now seriously. Both ELO and Elo are being used in publications. Elo, also seriously, is also a breed of dog.
I think that it is assholish to make an issue out of it.
As for the MC simulation - if you don't have stat background, it does not make sense to even think of why the simulations are good/bad thing. I have used it many times and the results are good.
@Harish Srinivasan: It sounds a bit unfair when you compare a new version of the currently strongest engine with old versions of its supposedly weaker rivals.
My Rybka 3.1 starts favoring 23.Sf6+ (!) at depth 13 (after 3:30 minutes on my 5-year-old laptop, 450 MB RAM).
The evaluations of both 23.Sf6+ and 23.Se5 is around 0.8 to 1.0, no big difference.
At depth 16 (after 24 minutes) it switches to 23.Sh6:+, just as your Firebird.
At the time of writing (after 1 hour) it still thinks about 23.Sh6:+ at depth 16, the evaluation is +1.38 and rising (slowly).
What eval does Firebird give?
If you think instances of incorrect use impress me, you are confusing me with people who count search-machine hits, poor fellows.
If you do want to impress me, quote the NYT instead of Xinhua.
"..
Huh?! You don't seem to have any concept of what Elo really is or measures. Topalov's rating is higher, so at the start of the match Elo would say, "Topalov has a greater chance of winning than does Anand," and not at all that "the match would go for Topa."
----------------------------
It is self-understood, silly!
Harish
Checked the analysis of the game on Firebird running on a simple dual core computer - it found the move 20. Nxh6 in about 2 mins - but the eval just gets stronger and stronger - started at 1.2 now upto 1.31 in 4 mins.
ELo/ ELO
http://en.wikipedia.org/wiki/Elo_rating_system
Just for some perspective for non-chess-players (and cricket lovers): today Vishy bowled like Akram's first spell of the Chennai test. Like Dravid, Topalov didn't do much wrong. Since Vishy is also known as the best tactical defender in the game - much like Sachin was on that day - the only way white could've been deprived today was if Vishy played black (too).
http://www.youtube.com/watch?v=KSAI8Q9e7XU
I found this comment kind of interesting.
In the interview for the Bulgarian National Television D-r Stefan Sergiev, the president of BCF praised highly the game, and gave his respect to the brilliant play of the world champion. There were also many positive reactions from Bulgarian players and fans, as Anand abandoned the passive positional approach and came back to his true active positional style. However, Krasi Kushev, candidate master and leading journalist from BNT reminded us that Topalov is usually playing better in the second part of the events, and we can wait for his come-back after the rest day tomorrow.
"ELo" is of course something entirely new.
But I promise to shut up now!
Come on guys, Nxh6 is extremely hard to find (or calculate) OTB. You will see the move but you won't find the move!! Black keeps material for a long time, that time would look like forever when you calculate all the arising variations and sub-variations OTB and that you would start to think if you really have any compensation for the sac. This is not blitz to expect your opponent to panic and to make some mistakes. A GM would need at least an hour or two, unless he goes in for some speculative sac. Speculative sac you can't do at the Wch level. I'm sure with a software running on single processor, even GMs may not see the win right away. This simply shows the preparation level and skill of Anand. When Topalov plays Rybka move like h6, it only shows he hasn't gone that deep in his preparation to see merit in the Nxh6 sac for white. It is so deep, yet Anand remembers, re-checks and executes with amazing precision just like a computer! when we all sit tight in awe of him!!
An amazing performance by reigning World Champion Anand!
@Bob: "As for the MC simulation - if you don't have stat background, it does not make sense to even think of why the simulations are good/bad thing."
Not sure whether that remark was directed to me... probably not. But just so that you know, I do have a Statistics/Probability background. Although that doesn't make my argument any more valid than the next guy, I have to say this: The fact that your model doesn't change the probability for Black vs White and doesn't consider the match situation make it a little too simplistic and hence unreliable. And just for the record, MC method is just a simulation method and no match to direct computation when it's possible. In our case, there are only 12 games each with only 3 possible results (not assuming another Toiletgate). So, at least for the classical portion of the match, you could consider all 3**12 i.e. 531441 possibilities and calculate the probability without going for any simulation. Anyway, I am not questioning your simulation method or whether it's good or bad. I am simply saying your basic assumptions are all too simplistic (and hence incorrect!) and thus irrespective of what method you use to calculate the final probability, it will be unreliable.
Anand is the Chennai Super King!!!
>Come on guys, Nxh6 is extremely hard to find (or >calculate) OTB. You will see the move but you won't find >the move!! Black keeps material for a long time,
agree. even though I did consider Nxh6 then e5 to get the
rook over the king side, i was not sure it would be correct. That was considered based purely on general scheme. However, not sure that it is 'extremely hard" for a strong GM (Topalov should have caught it). Still, point taken.
"I am not a chessgames Premium member, so I have to rely on someone in the Chessbomb chat quoting Short earlier in the game. From my memory, apparently he said something like "Anand again blundered a pawn in the Catalan, now he will exchange queens and win the game"."
Yes, the exact words were "So Anand has blundered a pawn again in the Catalan. All he needs to do now is to find a way to exchange queens and he will win easily..."
Most people got the (nice) joke, but of course as always some didn't.
Let us all pray that this rest day passes by without any statements etc from Mr Danailov.
Okay, it is hard to find or very hard to find. I think the key to the sac was, after e5 and e5xf6 to put the rook on d4 which is like giving a rook free (but black queen @ a7 can't take it because of Qg7#!) and tranfer it to g- or h-file to deliver mate. I couldn't see that idea so I didn't really know how to solve the Nxh6 puzzle. To see such a down the line idea OTB when you play Nxh6 is difficult. Rook transfer you will know but leaving a rook "en-prise" you may rule out. Also you will also have to calculate all the other variations like fxe5 with f7 pawn directly coming into attack from a7-Queen etc.etc. A small slip up or a hole in your calculations, you are doomed just like was in Game 1.
Let us all that all days pass without any statements etc from Mr Danailov.
You are referring to closed form solution versus simulation based solution. Well, very often close form solutions exist but it is very difficult to find these; particularly when combinatorics comes to play. The MS simulation will give an answer very close to the closed form (formula based) solution. When you increase the number of experiments, the two answers will be difficult to distinguish. About the assumptions, I agree it is always simplification .. but that's what we do when we have limited info. As a challenge, why don't you give us the formula to compute the probability? Why don't you do that assuming Topalov (win, draw, lose) = (0.23, 051, 0.26) - these are the probabilities prior to the beginning of the match. Besides, If you want to do it right, you would have to use conditional probabilities ala Prob(Anand wins|Anand plays with white pieces). If not, you end up with "simplistic" model. As I already pointed out, I did what was possible to be done from given limited info.
Give me the history of wins/draws with blacks/whites and then we'll talk again,
Hey, don't forget that I am not writing commercial package here.
I would appreciate it if somebody comes up with numbers. Then you'll see my responses .. :-)
Nah, don't think Danailov would do such things against Anand. I assume they have huge respect for Anand's play in general. If no cheating, then I'm fine. The only thing is, Anand should not lose focus and should score as much as he can with either color.
"but to be fair it's hard to reach across the board when you've just had your head shoved up your rectum like that." -Mig
Wow, and up to the Karpov endorsement I had always thought of Mig as neutral and apolitical.
More and more, seems he's a secret Anand-fan/Topalov-hater who's in conspiracy with Karpov-Kasparov-Carlsen to take over FIDE and Russia, j/k.
What cheating? Does it look like cheating? Also, these are things that have to be proven before throwing out. The losing side tends to accuse more often. Enough of this nonsense
Agree. I have noticed that people are critical being passive at the same time. If you don't like something, do it right and share the results. Otherwise = shut up!
Juan, I agree with Bob here - your critique is so general it could apply to any statistical model. Bob plausibly claims that his model accounts for the most important factors. You claim that accounting for the impact of the match situation in a realistic way would change the results. You need to show quantitatively that this might be the case.
You gave a fanciful example of how the match situation might change the probabilities, but you admit that example was unrealistic. Based on the empirical record of prior matches, what do you think would be a more realistic assumption for how the probabilities change as a function of the match situation? Bob can then plug those probabilities into his model, and I think you'll find they make little difference to the main result - after all, Topalov can try harder to force decisive games if he is behind, but this also means that he's probably reducing the number of points he'll get in expected value (since otherwise he'd be trying that hard to force decisive games all the time).
What is non-sense? Then why are they scanning the players? Why did they put the whatever cloth? Why are there anti-cheating measures? I'm not just refering to the Sofia championships alone. Any championship for that matter. What has to be proven?
"I believe that Anand's wins both result from outplaying Topalov over the board, while Topalov's win comes from home preparation, and possibly a memory lapse by Anand."
And I believe you are indulging in too much hero worship. Giri points out in Chessbase that Anand's Na3! was a strong move and the beginning of his preparation while Topalov was on his own from that point on. Looks like Anand is getting wins from good prep, just as he did against Kramnik.
We'll see if Topalov can come up with some prep of his own.
"ELO (not Elo) writing is the most often seen, read, that's why it feels as "naturally" the correct one.
"But if you want to be snobish(sic) you can have it as Elo and we will guess what you mean."
I wonder why Wikipedia was not consulted as is it usually is...oh, I see - the Wikipedia entry is for the "Elo rating system".
Chessbase : Elo rating
Oxford Companion to Chess : Elo rating
FIDE : Nothing in the handbook, both forms in various articles.
USCF : Elo-a-what?
@Toledo Paul: with respect, we will never know for sure will we whether it was prep or not - Anish Giri is hardly the authority on world championship prep by Anand - Anish is good but he has a VERY long way to go before we can take his word on Anand's prep.
to clarify - while Na3 was prep, we dont know where it ended and Anand seems to suggest at the press conference that Nxh6 was OTB.. that may be true (or not)- but we will never know..
I have read the analysis of the game from GM Giri, GM Alex Baburin (ChessToday), GM Illescas on icc and few others on the web.
Apparently there is no one common conclusion on what was Topalov's mistake. Now that itself suggest there is something special about this game and whites play. GM Giri think Rad8 is not the mistake as he thinks black is already lost by then. Only team Anand would know about the truth of the position.
On Arpad Elo... most use ELO. All caps gives notice that it is a different type of word. "Elo" is a name, but it has become more like "XEROX" which means "to copy" and not the company who makes copiers. ELO simply describes a rating system and not meant to refer to the author.
Not sure why people are saying Anand played like Topalov. No... he played like Anand. This is vintage Anand.
I'm wondering what Danailov will cook up to unsettle the Anand camp. Pressure is heavy on Topalov. However, he should not try to go all out to win Game #5 or he'll go two games down.
Of course every body prepares including Anish Giri. But I think it is important you make this distinction when you want to give credit to a player. I see two things. One is, coming straight out of preparation. The other is, coming out of the understanding that comes as a result of preparation.
Anand deserves credit for showing exceptional positional understanding that left even Top GMs clueless in Game 2. Not just directly out of memorized lines.
Anand deserves credit for his amazingly deep preparation and his calculation accuracy in Game 4.
Take this Game 4. Where is Na3 novelty and where is Nxf6? Na3 is 10th move and Nxf6 23rd move. That means, it is 26 level deep and you are prepared for a brilliant move. Nxf6 at 23rd move doesn't immediately show it is winning. You will have to go another some 5 moves or 10 level deep to realize it is winning. Anand is roughly playing at 36 level deep and sees a win. Topalov can deviate anywhere between 10th move and 23rd move and also a simple hole would just ruin everything. There is no obvious help from Topalov either on this game.
So this World Champion just stands out from all the former World Champions!!
Conspiracy rumours doing rounds in Bulgarian air waves: White bishops from both the sides are biased and prepared h1-a8 coronation route for Anand-Maran blacks in 2008 and now Catalan whites in 2010.
@Statistic: "Agree. I have noticed that people are critical being passive at the same time. If you don't like something, do it right and share the results. Otherwise = shut up!"
There is something inhererntly faulty in your logic. Let me explain. May be I am unable to "do it right" or unwilling to spend the enormous time it needs to "do it right." That doesn't mean I forfeit the right to argue it's correctness. Do you understand? If not, consider the case of an unsolved mathematical problem. Suppose mathematician X comes up with a faulty solution. If now mathematician Y points out a problem in the logic, would you ask Y to shut up since he couldn't "do it right" himself? Hmm... I guess I made my point. It's time for you to "shut up!"
I believe brilliance notwithstanding, Anand won game 4 because of his flexibility to throw conventional wisdom off the window. At high levels,Chess is more like Poker than a game of pure logic. The GMS play strong moves but send out multiple signals or keep the opponent guessing as to their overall plan/strategy. In most games, action on Catalan happens on the Q-side. Anand made Topalov believe that he's going top concentrate on the Q-side as conventional wisdom would have it by the Na3 move. Even the Ng4-Nd3 move indicated to Topalov that Anand's keeping his pieces close to the center to move them to the Q-side. After Topalov's Qa7,bam,the Knight goes back to Ng4 where it came from. In one simple move Anand transferred the fight to the K-side while Topa's hapless pieces on the Q-side watch the torture in silence.
Somehow reminds me of Kasparov's immortal game against Topalov some 12 years ago where Garry deftly maneuvers his Queen from the K-side to the Q-side on a Pircdefense where Topa castled on the Q-side!
In game 4, I don't know if Anand clearly saw a forced win when he played Nxf6, but I'd bet he saw a drawing resource before he took the plunge...
Re: Elo / ELO
As others have mentioned, all caps is typically reserved for acronyms. IBM (International Business Machines), UNICEF (originally United Nations International Children's Emergency Fund), are well-known examples. Using all caps for a word that is not an acronym is confusing, as it implies an underlying expression which doesn't actually exist.
In fact, I assumed for a long time that Elo ratings referred to some kind of three word formula (Extended Layered Ordination?) or organisation (European Labelling Organisation?), and wrote ELO myself, until I read a smilar thread on Arpad Elo and realised my mistake.
Of course, language is typically defined by usage, so if enough people write it in all caps, then eventually that could become correct usage, but the fact that it is not actually an acronym will probably act as a strong enough counterforce to prevent this from happening.
Anybody who wants to keep writing ELO can feel secure in being part of a significant minority, but why not simply adopt the correct usage once you know about it?
Pirc, yes Anand has worked out chess to a depth of "roughly" 36 moves. No one before him has ever played a novelty on move 10 and a winning sacrifice on move 23. As a lifelong member of the Anand fan club, I wonder if the secondhand embarrassment will ever end.
I said I would shut up, but I can't bear to leave this uncommented. Stone me, if you wish.
"Most say"... I shall start writing "definately", as that seems to be the present fashion. "Most say" is the most annoying criterion (this is the singular of criteria, by the way; it has long ago fallen by the wayside) for correct usage I know of. There are tangible differences between words used sloppily and words used as they should be.
"A different type of word"... Who has ever written Xerox in capitals?
And if the system no longer refers to its author, I am very sad that due attribution is considered irrelevant (do a search for that word, you'll see I spelled it incorrectly, if the majority rules, I suppose).
Consider yourself stoned. You probably were when you opined.
I am loving GM Giri's analysis.
" Black has stupid knights and a weak pawn on a5." Hehe, that's a kids' comment... but of course, a veeery strong kid indeed.
"So this World Champion just stands out from all the former World Champions!!"
Oh no! More Homerism and hero worship. Please spare us the hagiology. Are any of you Indian fans objective in any way? I mean, no one likes Vishy more than I do; I even gave my son the middle name of "Anand" (he was born in Thailand) in part due to my admiration for Vishy. But we are always reading drivel about Anand being "the greatest ever," or one of the most common: "most univesal champion," owing to winning the title in various formats.
As much as I admire Vishy, he was like a child compared to Kasparov. Gary tore him up most of the time, and their head to head record is terribly lopsided. So please spare me the "greatest of all champions" stuff. We are not kids.
btw, I do rank Vishy in my alltime top ten greatest players in history.
Is it still Anand - Topalov, or could it become Anand - Topailov (Topalov, Danailov, Cheparinov) if Topa really begins to TANK??
I'm sure the cheating duo Danailov/Topalov has prepared some non-chess moves for the situation they are experiencing right now. A new toilet scandal?
A good german site.
http://www.schach-welt.de/blog.html
Love the yoga comment, Mig. Ridiculously funny !!
I believe that Anand needs to go +3 (two more unanswered wins) to pass Topalov on the liverating list.
+2 would be enough to pass topalov..+4 to pass carlsen - which was a real possibility had tragedy not befell him in game 1..now he's likely to be a bit cautious with his black games..
Excellent analogy about playing styles. I vividly remember how akram was making the ball talk that Sunday morning but Sachin wasn't beaten even once in that spell. Great "tactical defence" by Sachin that day. It could also be said that Dravid's style is like that of kramnik's and he fell that day like kramnik did at Bonn...
Right... but the point is that ELO is not used as an acronym, nor is it contextually referring to Arpad Elo. When chess players say "ELO" (in caps)they are talking about a rating. If you ask someone, "What is your ELO?" this is not a question about Arpad Elo. It is a question about your chess strength. Yes... he created the system, but the context in which ELO is used is different.
Daaim: "When chess players say 'ELO' (in caps)they are talking about a rating."
Yes, in a way similar to saying, "I et a samwetch" when they are talking about consuming a hamburger on a bun.
An amazing contortion. I suppose you also refer to GALLUP polls? To NOBEL prizes?
Bob: "And, now seriously. Both ELO and Elo are being used in publications. Elo, also seriously, is also a breed of dog. I think that it is assholish to make an issue out of it."
And that's surely what people think of your making an issue of it in the beginning!
Danailov watch :)
Nothing too much to report yet. Here's one quote on Anand's tactics: http://topsport.ibox.bg/news/id_52761764
"We believe that in the coming games Anand will flee from theoretical clashes and with the black pieces he'll play catenaccio for a draw. He'll only keep looking for an advantage with white, as he's worse prepared than Vesko".
Kramnik's comments in Baku on Danailov perhaps trying some little trick have been widely reported in the Bulgarian press. I can't find it now, but Danailov was quoted as saying. "Since Elista [Kramnik] doesn't exist for me" :)
And Vishy Anand tore up the man (Kramnik) who tore up Kasparov, and has kept tearing him up everywhere he meets him on the board, most recently in Corus.
Look at Anand-Kramnik head-to-head in all formats.
So where does that leave Kasparov, Mr. Toledo Paul?
And that 1995 matchup between Anand-Kasparov you mention, Anand was pitted against the combined might of Kasparov-Kramnik (Kasparov's second), and Anand barely understood chess in 1995 the way he does now.
Also, Anand is spanning two generations - the pre and post-Rbyka.
>Those watching the video said that Topalov offered a very limp handshake upon resigning on move 32, but to be fair it's hard to reach across the board when you've just had your head shoved up your rectum like that. Takes a lot of yoga.
Mig, consider the possibility that you may have to pull that keyboard out of your rectum eventually. I can't believe you wrote that...
D.
Dimi,
-It all makes sense if you check out his day job on www.dailydirt.com
@.. When chess players say "ELO" (in caps)they are talking about a rating.
I suppose that for Knallo (who started this issue, not Bob) what is significant is who devised the system. He may be a distant relative of Elo or of the same nationality.
What makes you think I didn't get the joke? IMO it would have been an even nicer joke if Short hadn't said so, but someone else had invented this "Short quote".
The general point may be what we can and should expect from GM live commentary:
- jokes and entertainment (methinks Short and Susan Polgar are roughly in the same league), or
- trying to find out what's going on in the game. This failing, trying to find out what's going on in the players' heads (here Shipov is best, maybe it's futile trying to imitate his approach?)
Giri is somewhere in between, or has both. His notes are express rather than live analyses, so they may be a bit affected by the result of the game (but he has to start writing during the game, and can't make too substantial changes afterwards).
Maybe his most remarkable comment on game 4:
"23.Nxh6+!! Simple, yet beautiful! Vishy played it very quickly. For a player of his caliber, it's a piece of cake to find such a move! (Green with envy)"
Order has been restored - Giri is the student, Anand is the teacher ,:)
"Yes, in a way similar to saying, "I et a samwetch" when they are talking about consuming a hamburger on a bun."
Balls is making a joke that's going to go over the head of the recipient. Nice.
My view on the knight sacrifice: It isn't too difficult to _find_ and _consider_ it, nor to see the ingredients of the combination:
- the pawn cover around the black king is removed, and the white queen is "visiting"
- most black pieces are far away on the queenside, and cannot rush to the defense
- all it takes is a white rook joining the attack.
The difficult part is to calculate everything accurately, and to make sure that black doesn't have a hidden defense or counterattack.
A weak player may find and actually play 23.Nh6:+ without calculating till the end, and may actually also find the follow-up moves one by one.
An intermediate player might see and consider the knight sacrifice but shy away from it because
- he doesn't see the follow-up
- he takes too long to calculate the lines, and at some stage decides to give up ("I shouldn't spend 30 minutes thinking, if it doesn't work I will only end up in time trouble. Let's play something 'safer'.")
- he actually finds the follow-up but doesn't trust his calculations, maybe with psychology involved ("A player like Topalov doesn't miss such an attack, _I_ must be missing something").
And Anand is obviously a strong (and self-confident) player ... .
I assume during matches like these (and the tension between the two), Anand & Topalov don't do post mortems.
However, did Anand explain his Nxh6+, perhaps at the press conference?
@Giri is somewhere in between, or has both. His notes..
Giri Anish's notes seem very good to me, or I can't find many which bother to go in depth instead of few wordings added to the moves we all know. It would be good if people would post here the links the know to the sites where we can see in-depth analyses.
see his
17..Nd3!? sac (instead of 17..Nba6) 18.b3 Nxf2 19.Qxf2 cxb3 20.Rd3 bxa4, with complications
and
20..Qe7 (20..h6?) 21.Nxa5 Bb5 22.Nxa5 Rfd8 23.Nd6 Rab8-and black has (sort of) solved the development problem by returning the pawn
Since Hertz, Newton, Joule, Pascal, Celsius, Volta, Faraday, Watt, Ohm, Tesla, Ampère, Kelvin, etc have the fundamental units bearing their names written with only the first letter capitalized, why would it be any different for Arpad Elo? Nobody ever makes measurement in HERTZ, NEWTON, JOULE, PASCAL, CELSIUS, VOLTA, FARADAY, WATT, OHM, TESLA, AMPÈRE, or KELVIN. Oh, and don't scream, it's considered impolite on internet.
"why would it be any different ?"
Because it IS different.
By your twisted logic "Watt-watts", "Newton-newtons", etc.. the rating points would be "Elo-elos".It would really weird to say :
" Carlsen has 2813 elos",
correct is :
"the ELO of Carlsen is 2813"
Seriously, guys, I have 400 Elo points less than Anand, and I'm flicking to the game between work, and even I saw up to 26 exf6 and saw that I couldn't see a move for Black. If this were a white-to-play-and-win problem, I assure you that any 2700 player would see the move and understand that it was at least very dangerous within 60 seconds.
Anand didn't see it when Topalov played 20...Rad8. He started thinking about when 20...h6 was played. That's the difference.
As to why Topalov didn't see it, who knows? He does miss opponent's tactics under pressure; that's what let him down in Elista. But I'm absolutely sure he'd have seen it if he'd been White.
By contrast, Anand's 16 Ng4 didn't occur to me; that was far more difficult IMHO.
$10 still on the table folks...find me a toiletgatesque story!
Anyway, the more important news is that the Habs beat the Caps in Game 7 of the NHL playoffs.
Then you have some work to do, Ovidiu. By all means, rewrite the entire wikipedia page on the Elo rating system, because not a single instance of it is written ELO. Oh, and while you're at it, you might want to remove from the corresponding French page the obviously erroneous sentence:
"Elo se trouve parfois écrit par erreur en haut de casse ELO. Or, il ne s’agit pas d’un acronyme. Il doit son nom à Arpad Elo (1903-1992)"
(Elo is sometimes wrongly written in all caps ELO. It is however not an acronym; it owes its name to Arpad Elo (1903-1992)")
@ Ovidiu. By all means, rewrite the entire..
I did not care how you wrote it.
It was you (and Knallo) who began this silly subject by starting to "correct" people here (Bob, Daaim).
Great to hear comments from people with a rating of 2300+. They can actually comment sensibly on games for the benefit of the rest of us. They also have a lot more respect for the players because they have experienced the ups and downs of competitive play themselves.
I'm always amazed at the huge number of 1500 rated muppets that crowd on to sites like this, making retarded comments like "Topalov can't play positional chess". So a 2800+ player can't play a subtle positional move? Right.
How many of you took up chess a few months ago and feel you can ridicule the best players to ever play chess just because you bought a copy of Fritz and it found a few moves for you? Your analysis skills are so ridiculously feeble that without your computer you see absolutely nothing. And because you rely on a computer to work out tactics, you will never progress.
Have a bit of respect for your betters ffs. Even Susan Polgar would kill all of you in any format 100 times out of 100 and you wouldn't even know where you went wrong.
I certainly did not started this nonsense. My first post on the subject is the one about Newton, Kelvin, etc dated April 29, 2010 5:20 AM. I was hoping to put an end to it. Obviously, I should have realized you were just trolling. Fair enough. Write ELO if that makes you happy.
@Fair enough. Write ELO if that makes you happy.
many thanks, Guillaume
What makes you think I thought you didn't get the joke?
"If this were a white-to-play-and-win problem ..."
Then it would indeed be rather easy - once you KNOW that white can win by force (because you were told), 23.Nh6:+ is the only candidate move, or at least the most obvious one (apparently some engines suggested 23.Nf6+).
However, Anand didn't have that information, at most he could suspect that white MIGHT have a forced win - the knight sacrifice is tempting, but (until you calculate the lines) it may or may not be correct.
BTW - not saying that it happened - here some non-verbal cheating could have been beneficial: some sign from the audience meaning "go for it" or "THINK HARD!", anything clarifying the difference between "there may be a forced win" and "there is a forced win". If roles had been reversed, it would have been a strange coincidence if a door alarm had gone off just before _Topalov_ sacrifices a knight ... .
BTW, I don't understand why 16.Ng4 was hard to find or surprising for you: the knight was under attack on e5, other candidate moves were the piece sacrifice 16.dc6:, the loosening 16.f4 or the 'strange' 16.Nf3 (blocking the Catalan bishop, and where does the knight go from here? g5?). So for me (ELO/Elo around 1950), Ng4 was the most obvious move.
"retarded comments like "Topalov can't play positional chess". "
I don't know if anyone made such a strong and categoric claim, I dare to make the following claims or statements:
- Topalov doesn't like to and doesn't feel comfortable playing positional chess
- Topalov is _relatively_ weak at positional chess, compared to both himself when he gets an attacking position, and to Anand (or Kramnik) in quieter waters.
Both actually doesn't refer to positional chess in general, but to what Mig calls "passive groveling" or what I may name "the art of manoevering and doing nothing (special)". I couldn't imagine Topalov behind the black pieces in game 2 ... . Accepting that a game may end in a draw is also a bit against the spirit of Sofia rules, thus against Topalov's (and/or Danailov's) own commitment.
GM Christian Bauer and IM Alexander Ipatov at chessdom.com :
15... Qd6
A lot of options now for White here 16.Qc3, Qd4, dxc6... not easy to decide between taking back the pawn, with an approximate level game, maybe a tiny plus to White. Two ways for white are interesting 16.Ng4 (a calm one) and 16.dxc6 (a very complicated one)
Yes, fair point about white-to-play-and-win; OK, let's say a problem position with the question 'evaluate 23 Nxh6+'.
I was thinking mainly of queen moves or else dxc6 at move 16. That's because I'm weak, of course, but still I think finding where among several promising paths White's chances lay somewhere between moves 15 and 20 was the hard part of this game, not the knight sacrifice. The times suggest that, as well.
Bob: "...the personal history between the two players is more important than the ELOs."
Statistically speaking, the answer is "it depends."
A standard measure of how good an estimator (say, T_hat to estimate T = probability of Topalov winning) is the mean squared error, MSE = Var(T_hat) + bias(T_hat)^2. Picking games just between Topalov and Anand, you reduce the bias, but because there are so few relevant data points (15 classical games in the past five years), the variance is extremely high. On the other hand, if you look at all games played by players with Elo > 2750, you introduce some bias, but the variance in the estimator will be reduced substantially and you'll wind up with a lower MSE, i.e. a more reliable estimator.
The numbers for Elo 2750+ GMs in classical time controls are approx P(W wins) = 0.27, P(draw) = 0.60 and P(B wins) = 0.13. Color matters in a MC simulation, especially when there are an odd number of games remaining. Running these numbers at this point in the match, I get P(Topalov wins match before any tie-breaks) = 0.22.
That was the mother of all smackdowns. Almost clubfoot-esque in brutality.
I like your 'analysis' of player thinking processes based on strength.Thank you.
I would instantly think about the Knight sac because I always think of spectacular moves and because the positional considerations suggest i should attack in the sector where i was stronger. However, I might not make the sac because I could neither work out a forced a win nor a way to keep the draw in hand.
I think it's this kind of enhanced thinking, recognizing efficient patterns, that marks out stronger players. And it is a model for aspiring players to follow.
Stockfish 1.71 gives me Nxh6 in just over a minute. I believe that it is not all preparation as Ian Rogers on his USCF write up mentioned.
Anand may have seen this structure in prep, but I think this is more likely OTB. He made the sacrifice after assessing the position with the best black can do for a defense.
What everyone is missing out on in saying he mad his moves quickly is that Anand is very quick with this calculations. As his game matured he has used patience and speed jusiciously. I do not think the speed of evaluating the position has slowed any.
Couple of corrections to my post
Stockfish 1.71 gives me Nxh6 in just over a minute. I believe that it is not all preparation as Ian Rogers on his USCF write up mentioned.
Anand may have seen this structure in prep, but I think this is more likely OTB. He made the sacrifice after assessing the position with the best black can do for a defense.
What everyone is missing out on in saying he mad his moves quickly is that Anand is very quick with this calculations. As his game matured he has used patience and speed judiciously. I do not think his speed of evaluating the position has slowed any.
A question: Since it is obvious how much prep goes into the World Championship matches, do you think the holder and the challenger have any significant chess life left apart from (playing and preparing for) the championship matches?
A question: Since it is obvious how much prep goes into the World Championship matches, do you guys think the holder and the challenger have any significant chess life left apart from (playing and preparing for) the championship matches?
Thx senthil for the Ian Rogers-link. Geesh, it seems that there as many opinions as players, regardless whether are GM's or less than 2300(ELO of course).
"Only when I saw the variation 27...Bd5 28.Rc4!! Bxc4 29.Rd4! was I sure that I was winning," Anand admitted - though he didn't mention whether he first saw this variation at home or during the game.(I.R.)
leon lederman once said that this was the difference between university intelligence and scientific intelligence: in university, you know there is an answer at the back of the textbook, the problem has a solution that is within the reach of the material you have been taught. that is not the case in real science, where choosing the interesting problem that you can solve is extremely important.
@ I believe that it is not all preparation as Ian Rogers ..
only Anand (and his seconds) know for sure, up to 14.Rfd1 and the 15.d5 break he played fast but after 15..Qd6 he used a lot of time for 16.Ng4 and 17.Ne3
I note that in the game 4 press conference, Topolov once again asserts that the Sophia rules govern the match. Apart from the continued obnoxiousness in that regard, I thought Topolov looked more comfortable after this loss than he did after the previous day's draw, and I doubt that the game was any more of a psychological setback for him than Game 1 was for Anand.
Given that the Team Toilet has painted itself into a corner with respect to draws, Vishy should strive for another awkward draw tomorrow of the type where Topolov can't do anything without hurting himself but also can't offer a draw. A position less prone to repetition would be ideal. Although Team Toilet labors under the assumption that Topolov has more physical strength and endurance than Anand, I'd bet the opposite is true with respect to nerves, particularly over the long haul.
"Find the best move for white" may be even better, to check whether one even considers the knight sacrifice.
Earlier on move 16, I (obviously) didn't at all think about queen moves - I would say this isn't related to playing strength but maybe to "style" or rather how one usually handles the different pieces. As a remotely comparable example: at all levels, some players generally prefer bishops over knights, some rather prefer horses over what Russians call elephants ... . Among all annotators, the Chessdom guys seem to be the only ones suggesting queen moves - several others say Anand had to choose between Ng4 and dc6: (by itself justifying a 22-minute think).
I agree with you that moves 15-20 were the critical phase of the game - if white isn't careful and precise, the compensation for the pawn may evaporate and he ends up worse? However, it remains unclear if, how and how quickly white would have won if Topalov didn't play the careless 20.-h6?!
It's quite sad to see that nobody knows what part of an exceptional game belongs to the player and what part belongs to the computer. It is generally assumed that the computer is the one who found the winning idea(s).
And yes, there has been a degree of home preparation since organized chess came into existence. The difference is that in the pre-computer era, the moves were found by the players themselves (often with the help of "seconds"). Nowadays, the computers not only crunch the numbers, they also provide the "creativity"... relegating our top players to little more than chess parrots, real pawns in the real match betweeen their respective computers.
I find your "relegating our top players to little more than chess parrots, real pawns in the real match betweeen their respective computers" not only facile and uninformed but a pathetic attempt to denigrate player abilities. Shameful.
The problem with your argument is this:
Before computers, the level of play was equalised by the two opponents doing home prep with their seconds. Both did it so one did not have advantage over another that he would have if he did not do home prep.
On match day both opponents came to the board with equally an home prep effort, so OTB their abilities came into play unless player 1 'blundered' into player 2 home prep and the player 2 was lucky not to blunder into player 1 home prep.
Nothing has changed with the advent of computers.
Now, both have computers, neutralising any advantage and with omputer power the element of a surprise home prep is much less than it was there before.
That is the reason that players like Topalov who rely more on home preparations cannot deal as effectively when questioned in positional games OTB.
Moreover, computer helped home prep of say, 26 level deep mean nothing if the opponent deviates in one of the moves. Then what?
>Nowadays, the computers not only crunch the numbers, they also provide the "creativity"...
Nonsense. You have no idea what you are talking about. Maybe you should try finding a worthwhile opening innovation yourself with the aid of a computer?
I haven't understood this 27...Bd5 28 Rc4!! variation. Why doesn't 28 Rd4 at once do the trick also, I wonder?
@ relegating our top players to little more than chess parrots..
well, not quite "parrots", the become "entertainers" who come to play on the stage their (memorized at home) script. It is all for us to see, comment and enjoy .
Ian Rogers has it very well :
"Just as at Wimbledon, a big first serve is crucial and the rallies will look after themselves. Yet if the remaining games are half as entertaining as the first four, who cares?"
>>> Nowadays, the computers not only crunch the numbers, they also provide the "creativity"
Haha. Talked like a true idiot. If only it was that simple as you suggest you would have found yourelf playing chess instead of running down brilliant calculating minds playing chess.
It is like saying if enough monkeys were given paint and brush they would eventually paint a picasso :-)
Oh, the rook's on d6 of course, not d1. D'oh!
R at d6, B at d5, Rd4 not possible
I bet Ovidiu would'nt be saying this if his "chess parrot" had been winning the games Anand did.
Still, like a fly drawn inexorably to the flame, the troll Ovidiu hovers on daily dirt all his waking day and sleeping night even :-)
The only time Ovidiu throws away his "troll mask" is when his beloved Topalov is winning. Then he i all about Topalov's brilliance OTB.
Hail the troll, nee the joker :-) He just cannot stomach that Anand can beat OTB all of Topalov's home prep.
The Bishop is in the way because the Rook captured d6.
It's Rc1-c4, Bd5xc4, then Rd6-d4.
@then he is all about Topalov's brilliance OTB.
don't surpass yourself chess_tact, Rogers' article aims at both players :
Much was made before the match of Anand's understanding versus Topalov's determination and attacking flair but at the moment it seems what really counts in this world title contests is deep opening preparation against your opponent's favorite systems and the technique to finish off the game after you have achieved an advantage.
Ovidiu,
I'm inclined to agree with you on when the prep likely ended ended.
As I noted earlier I.R is favoring an all prep conclusion which I feel is not warranted.
Anand has the lead? Look out, here comes a Danailov instigated scandal. Topa mustn't be disturbed so D will do the dirty work of smearing Anand. So the question is, what defensive strategy does Vishy have worked out for this?
Anand believes that deciding the world champion in a tournament is reasonable. In a way that may be a solution to move away from such deep computer preparation.
Put top 10 players in a double round robin and decide who gets the best score. There is still going to be computer preparation. However, trying to prepare for and against 9 other players with white and black is going to force players to bring their inherent skills as opposed silicon dictated moves!
Just my 2 cents.
Anand believes that deciding the world champion in a tournament is reasonable. In a way that may be a solution to move away from such deep computer preparation.
Put top 10 players in a double round robin and decide who gets the best score. There is still going to be computer preparation. However, trying to prepare for and against 9 other players with white and black is going to force players to bring their inherent skills as opposed silicon dictated moves!
Just my 2 cents.
Ovidiu - I find your observation weak on several counts.
1. GM Ian Rogers must be a jerk to base his observation only on the fact that "This must be home prep since he played fast". What a jerk. So GMs who might've calculated OTB cannot play fast? Or he cannot play fast based on his understanding of the situation OTB from years of playing top level chess.
2. For a 40 move game I am assuming that a player has to chart a 80 move deep game (40 moves x 2 players).
The permutations and combinations must run into their hundreds if not thousands in any line arising out of say, a 5 move advance into the game (am hoping the knowledgable here will help shed light on the actual permutations and combinations possible in leading to an ending line at any given point in time in the game).
3. Each of the players is aware that the other will have prepared some lines and will hence try to negate the line with their own 'prep' as it were, in which case the other player must necessarily be prepared to counter it OTB.
4. In Anand's game 4 there was talk about a novelty he played on move 10 (if I remember correctly) and then another 13 moves later at move 23 (if I am correct) which would mean that at move 10 Anand should have calculated at 26 level deep (13x2) to gain an advantage.
26 level deep calculation when at any point the opponent could have interrupted the calculation!!! Must be sheer genius at calculating.
5. Could it be possible that super GMs have a better pattern recognition ability and an intution of which line might pay better dividends.
6. Both players use computers so computers have enabled both computers equally, it's not as if one is using a early prototype of Fritz and the other the latest version of Rybka. So they are evenly matched on the silicon front, thus testing their actual abilities OTB.
Sure some games will be about one player walking into a home prep trap. It was the same in the olden days that Ian Rogers nostalgically talks about as well.
Still, players have to outmanuvre the other on the board, struggle for advantage, feint the opponent, outpsych with moves, consolidate the advantage after gaining it and execute the endgame perfectly because there's the small matter that opposite you is a player attempting to deny you these very things and is trying damn hard at that.
By the way jerks like Ian Rogers are good for the advertisement of chess as a top level sport. Or is it a case of sour grapes for never having made it to the top despite having access to the computers that Ian Rogers talks about playing each other OTB unless ofcourse he means they have computer-like abilities.
Parim,
Why do you resort to name calling a GM who has probably more experience than all of us put together.
IR is merely taken the argument to an extreme point and forcing us to view what might be lost in an era of computer based prep.
All of your points are valid. On this particular topic Ovidiu and I are on the same page that IR's conclusion may not be entirely correct.
People thinking that Nxh6 was a home preparation simply because Anand played it in 4/5 minutes are just ignoring the simple knowledge that all expert Chess players (I would say most of the GMs, some 800-1000 of them) would find it.
Finding the move Nxh6 is not so difficult. Any human Chess player would intuitively see that move. It is such a well known pattern. Black King trapped on g8, White Q covering the h-file and White has number of squares for Rook Lift and move to g-file.
It's actually a difficult move for computers because they have to go for concrete (and therefore very deep) variations before there evaluation compensates for +3 score of the Knight sacrificed. Humans don't need a concrete score. Lot of GMs who play attacking chess would go for such a line gives them initiative or at least a draw in case, things don't work out well.
What are the exact Sofia draw rules? You can only offer a draw through an arbiter if the position is a technical draw?
I think Anand toyed with him on Game 3 by not offering a draw via arbiter and kept playing forcing Topa to request a draw after a 3-move repitition. In the scuttle, they forgot to shake hands and Anand joking said in the press conference that he thought only the arbiter can shake the hands!
He's clearly making a point :-)
I don't think anyone really believes Anand had the Nxh6 sacrifice in his analysis (although certainly similar sacrifices might have featured, with different actual positions). He himself said he wasn't sure he was winning until after he calculated the double rook sacrifice line.
Maybe what Rogers meant was "What else was Topalov going to do?" At move 20, White appears to have a substantial advantage. 20...h6 was meant to enable 21...Rad8 (delayed by a tempo on the queen in the game). Both of Black's a-pawns are weak, and the pieces on the Queenside are no longer doing very much. Given time, White will play his knight to d6 and mount a kingside attack.
To stop this Black needs to contest the central files (hence ...h6 planning ...Rad8). Without d6, White's knights may trip over each other and if White playes e5, the d5 square becaomes available again.
White is actually doing very well even before the sacrifice.
Senthil,
I said Ian Rogers is a jerk to conclude so sweepingly that it is 'computer versus computer' precisely because he is a GM and knows better than all of us.
It is because he is a GM and said that makes him a jerk because he should have known better than to say it.
Parim, I do not fully agree myself with Rogers.
I pointed out that, based on the "time-analysis" , Anand's home prep seem to have ended after 15.d5 when Topa played 15..Qd6. That was not in plan, it got Anand thinking some 40 min for playing 16.Ng4 and 17.Ne3.
If that's not so (if Rogers is right) we would also have to conclude that Anand was playing a theater to delude Topalov.
But Rogers has a broad conclusion :
" What really counts in this world title contests is deep opening preparation against your opponent's favorite systems and the technique to finish off the game after you have achieved an advantage"
I think that we should keep it in mind to see if the match (as a whole) will confirm it as being actually the point, the key variable (as oppossed to others, say : age, ELO, or the personal score in classical games) about these modern WCC.
Anand didn't offer the draw in Game 3 because he was Black, and the weaker side in the position. It would be rude to offer a draw when your opponent has the "better half" of the draw. You should play on, and wait for your opponent to offer the draw, as Anand did, because your opponent has a perfect right to test you in the position until they decide it isn't worth it.
The weaker side offering draws is a waste of time.
Top level players don't trust their prep. If they suddenly get "interested" in the position, they will absolutely start thinking before they are out of preparation, and make a different move if they believe it is warranted. Anand personally has said things like that before in his notes.
So the mere fact that he spent 40 minutes on moves 15-16 doesn't -necessarily- mean he was out of prep at that moment. I would be shocked if it did, since ...Qd6 is a computer favorite in the position and cannot possibly have been a surprise. He was probably checking his memory and possibly deciding -between- different lines he had prepared (I doubt they prepared dxc6, so this probably means Ng4 and Qc3).
If so, maybe even the sacrifice was prep, but clearly the follow up was not. Probably Kasimjanov knows those lines by heart, but just told Anand that ...Rad8 is bad because of Nxh6 - and figured Anand could find the rest.
Could be, in the first game Anand played 23..Kf7? (24.Nxf6 !) instead of the 23..Bd7 after 26 minutes of thinking, he may have got "interested in the position" after 23.Rf3, as you said.
Well said Bartleby!
Dude, you are my hero.
In this exchange on human creativity vs. computer dictated influence on chess play, there are broader principals that should be recognized:
I would like to take IR example of Tennis and Wimbledon:
When the racquets went from Wooden to Graphite based and increased in size, there were lots of doomsday scenarios raised about how Tennis would be robbed of it's rallies and about how technology will supercede skill (I do not recall the source, but one writer took an extreme stance on the success that Michael Chang had attributing it more to technology than skill).
If as IR suggests that the serve is all that may matter, then Andy Roddick and not Roger Federer / Rafael Nadal would be the topic of conversation. It is interesting that neither RF or RN have the fastest serves in the game, just very good mastery of the game. Goran Ivanisevic, a favorite of mine could only win one Wimbledon.
I feel that material used in racquet construction is only secondary to player's skills and the surface since all players have access to the same type of racquets.
In a similar vein, it is only a natural progression of how technology is intertwined with chess that all players (a 1500 to 2800 level player) can now have access to excellent software to study and prepare in chess. However their skill, preparation, patience, and temperament play a crucial role in their ultimate success. In yesteryears among the Russina players, they had access to chess databases, training, and structure that players in other countries could only dream. Today with Deep Aquarium and if I pay $110 then I have access to about 4 million games.
If anything computers and computer software has leveled the playing field.
Ultimately it will come down to OTB play. Hence introducting variations to the game may be an answer: Fisher Random Chess is one such clever invention.
Rather than rue the influence of technology on a sport/game we should give credit to human genius in not only creating a game or a sport but also in adapting it and making it evolve. There in lies a constructive challenge to the human mind.
Why are you calling other commenters petty names? Especially when the player you seem to support is doing well? This is the time to be gracious, I would have thought.
Just one more thing about the cheating problem in modern top-level chess. Most people seem to think that signalling moves is necessarily a matter of one-way communication.
How about this:
A player is out of book and gets a critical position or he simply forgot the excact move order of his preparation. There are three candidate moves, one with his king, one with his bishop, and one with his rook. He thinks for about 15-20 minutes but he cannot decide which one is best. So he gently rubs his nose (= shall I move my king?) and after a few seconds there is an inconspicuous signal meaning "yes" (someone yawning in the audience, a light that briefly starts to flicker, a small shadow that suddenly appears on a particular spot of the wall, etc. etc.). If there is no signal after 30 seconds, he briefly rubs his left ear (= shall I move my bishop?) ... you get the idea.
In many cases just asking which piece he should move would be enough. If necessary, you could also refine this (e.g., "rub left eye" and then "put both hands briefly under the table" = shall I move my rook to the c-file").
If you do this only three or four times in a game, it is practically impossible to detect.
Obviously, you cannot turn a complete patzer into a world champion this way, but it could make all the difference between 2700 and 2800 or between "5.5 : 6.5" and "6.5 : 5.5".
Think of what difference it could have made if Anand had used such a code just once in game 1 (before 23 ... Kf7). He may well have drawn or even won the game.
Again, I do not accuse anyone of cheating, but if you fail to see the problem here, you must be naive or an utter philanthropist. The situation is particularly problematic if a match or tournament takes place on the home turf of one of the players and/or if there is no clear distinction between the organizing committee and the team of the local player.
I predict that these things will eventually be the death of top-level chess at classical time controls. Perhaps rapid tournaments (say 20 - 25 + 5) in cubes made of one-way mirrors, sound-proof, blocking all radio and other signals, are a viable replacement. After all, you can reasonably expect a player to stay at the board for up to an hour.
Amen,
kuk
I have been thinking about IR lamenting the good ol' days when preparation didn't rule, something which as far as I know, Fisher was the first big player to protest about. But Fisher himself 'prepared' a lot of variations, even Marshall as early as 1920 prepared his gambit specifically for Capa and waited for 8 years to spring it on him! Of course famously Capa deconstructed the prepared TN OTB. So, players preparing for specific opps specific novelties is nothing new. Ok, previously the players did it on their own, then took the help of seconds and have now added computer to the fray. But still it is the players who have to choose the openings they want to innovate upon, and can use the computer only to assist them in the task. Sure, computers might find decisively the new moves if left to run long enough but that is ground reality now. Lamenting about that is as pointless as lamenting that tennis rackets these days have gotten to be technically better and better and give the players some starting advantage that wasn't their before.
To what extent a computer can assist a player depends on the LEVEL of the player. It is not that all of us can use the same computers and come up with the same analysis as Anand. The kind of moves/variations he (and his team) would test is different from what a player at a lower level would and can test.
In this instance, I played through Pein's revised comments on Game-4 and one can see how superbly the play in the game ends in Anand's favor against so many different responses after 23.Nxh6. So the credit has to go to the player bcs without deep understanding this player could not have come up with this preparation.
There is a special (and very expensive) curtain in place in Sofia so that the players cannot see the audience. This is precisely to avoid the sort of signalling Kuk suggests.
It is self-evident that the whole "it's all computer prep" argument is completely senseless. Computers have been in serious use for at least 15 years. If they were very important, they would serve as the great equalizer. Everyone should end up playing at a similar level, because they have access to the same computers and databases. There would be no distinction between say a 2600 player and a 2700 player, it would just be a matter of who walks into whos prep. The fact that there is a clear hierarchy of players proves that there is no equalizing force at any level: it boils down to chess ability, like it always has.
It is funny the length some people will go to in order to not give someone credit for their play. When Anand won the WC title through a tournament (which was declared as such beforehand), it was "but it's a tournament, so it doesn't count. REAL WC are decided by match play only." Then he won through match play, and now match play doesn't count (unless, of course, his opponent wins) because it is all prep. LOL.
dee4,
In fact playing via a double round-robin tournament will make it very difficult for any player to simply rely mostly on preparation.
I never thought about it in this manner even when Anand felt that via tournament was as good in it's own right. From a very unlikely point of view I now feel this can be potential solution
The advantage of tournaments is that it does not allow for deep preparations against a particular player. But the big disadvantage is that whether you win or not is determined partly by games between other players. You are not in control of your destiny (unless you win every single game, which is only possible in theory).
First, there is a clear possibility of collusion. You can support player A (by offering easy draws or throwing games) just to make sure that B does not win.
Even if there is no clear collusion, things can get unfair. Say player A badly wanted win, and prepared his heart out, but by the time last one or two rounds are played, it is clear that he has no chance of winning. He is all heartbroken. He does not put his full effort and loses to player B or offers and easy draw, who needed just that to beat C and win. But when C and A played, it was early round A had to face C's full force. In effect, C does not win because A decided to give up against B. That's why deciding WC by tournaments is problematic, and susceptible to all kinds of read and imaginary conspiracies.
Another knock on tournaments is that you can become a WC without beating the top 1 or players. All draws with the top players, and a few wins against the weakest players could make you a WC. Many would find it unsatisfying to have a WC who has never beaten his main rivals.
Another knock against tournaments is that you can become a WC without beating the top 1 or 2 players. All draws with the top players, and a few wins against the weakest players could make you a WC. Many would find it unsatisfying to have a WC who has never beaten his main rivals.
Dude, you see toilets everywhere. Do you spend much time in these?
That you and everyone else knows with 100% certainly to whom "Team Toilet" refers is a fact that speaks for itself. I'm hardly responsible for that legacy.
Well said!
No I don't, but I refer to FIDE ratings as ELO. Gallup and Noble are not euphemisms for another concept. ELO (which means rating) is a unique case. If I ask you "What is your ELO?" what will you say?
Didn't mean euphemism... more like a simple substitute. Anyway... who cares which is used. I believe chess players know that one's ELO/Elo/elo/ means rating.
After threefold repetition, couldn't Anand claim a draw instead of waiting for Topalov to offer a draw?
anand going up 2.5-1.5 is a nice way of replying to the partial and biased organizers and to tapalov's manager who is so rude & unsporting...FIDE sud think twice b4 holding any championship of this stature in bulgaria from now on....
anand going up 2.5-1.5 is a nice way of replying to the partial and biased organizers and to tapalov's manager who is so rude & unsporting...FIDE sud think twice b4 holding any championship of this stature in bulgaria from now on....
"There is a special (and very expensive) curtain in place in Sofia so that the players cannot see the audience."
-----------
This kind of curtain was first used in the Bonn match between Anand and Kramnik in 2008. Maybe as a consequence of all the hoopla in Elista in 2006?
BTW: I don't think the curtain is that "special". It is probably a traditional voile ("see through curtain") made from ordinary 100% Trevira CS Polyester fiber with a certain density, sewed with a led bottom hem to keep still. The mirror effect is created simply by diming the light on the audience side.
This blog has been refreshing for all those who considered that only Bulgarians posted distasteful, borderline insane, conspiracy stuff on behalf of their hero.
We now see that plenty of Indians do that too. Can it be that every country has crazy people, just like Dimi always said?!
The other thing that comes over strongly is what a shame it is that Danailov is such an unsatisfactory personality. If he had class to go with his evident abilities, what a force for good he could have been.
Yah rdh but India is 1.2 billion people, while Bulgaria is what, 8 million? The crazy ratio race is totally won by the Bulgars ;-)
No doubt every country, race, creed, class, culture, religion and any other denomination has its fair share of jerks, loonies etc. No exceptions.